I think it's dying

[papaschtroumpf]

Yesterday my data drive on my windows machine disappeared. Windows doesn't list it in Computer Management (win7) and the Intel utility that came with the Mobo flashed a message it was disconnected.
I had just performed a giant Windows update which included some motheroard and choose drivers, so my first thought was this was the cause.
I rebooted in the BIOS and the disk was not detected at all (as it the port was empty), booting to Windows, the drive was there, then disappeared 15 minutes later while I was trying to back it up.
Booted back in the BIOS, and it showed up there (SATA port 0), booted to Windows and it seems to remain accessible for the last couple of hours.

Since it was not visible in the BIOS I don't think I can blames Windows updates?

My guess is the drive is going bad? What else could it be? SATA port on the Mobo (do those go bad?) or maybe that output on the power supply? (Everything else seems fine with power).

Is there a better way to determine? I have SMART enabled in BIOS but not sure what that's supposed to do. The Intel utility is not reporting anything either.

 

[UsandThem]

You can try one of these utilities under 'best answer':

http://www.tomshardware.com/forum/282651-32-best-diagnostic-testing-utility

And you can also run chkdsk in Windows:

http://www.howtogeek.com/howto/windows-vista/guide-to-using-check-disk-in-windows-vista/

 

[papaschtroumpf]

The Intel storage utility that shipped with my motherboard is supposed to monitor SMART on the disk and never said boo. It only let me know that the drive is being disconnected.
The fact that I don't get SMART errors is what is puzzling me and makes me doubt the problem is the hard drive itself.

I'll check utilities on that list and I'll swap power, SATA cable, etc... also

 

[tortillasoup]

Weren't there some Intel Motherboard chipsets that were vulnerable to actual hardware failure? What motherboard and chipset do you have?

 

[papaschtroumpf]

Not at home to make sure, but pretty sure the motherboard is ASRock Z77 Pro3 ATX Intel Motherboard, so a Z77 chipset

 

[tortillasoup]

Try to back up the drive on a different SATA port, instead of a black port, try a grey port. The cable could also be bad so try swapping it with the one you boot the OS with or something.


Also when you're in windows, try slightly bending/flexing the drive cable and hold it, see if it reappears. Could be a worn out port or slightly damaged cable/SATA socket.

 

[papaschtroumpf]

CrystalDiskInfo reports everything nominal except:

05 100 100 _36 000000000028 Reallocated Sectors Count
C5 100 100 __0 000000000048 Current Pending Sector Count
C6 100 100 __0 000000000048 Uncorrectable Sector Count

I don't really know how to interpret SMART values but the Acronis database says all 3 are critical and could be a sign of imminent failure. Sounds like I should replace the drive?

I moved it to another port and so far so good, but since the problem has been intermittent it's hard to know if that was the issue.
If I get a new drive, I assume it's safe to clone this one into the new one?

I looked up the drive and it is a Seagate Barracuda ST3000DM001 3TB 64MB. Funny enough I paid $100 for it in 2012, and its updated version is still $100 on newegg today (ST3000DM008)

 

[Elixer]

CrystalDiskInfo reports everything nominal except:

05 100 100 _36 000000000028 Reallocated Sectors Count
C5 100 100 __0 000000000048 Current Pending Sector Count
C6 100 100 __0 000000000048 Uncorrectable Sector Count

I don't really know how to interpret SMART values but the Acronis database says all 3 are critical and could be a sign of imminent failure. Sounds like I should replace the drive?

I moved it to another port and so far so good, but since the problem has been intermittent it's hard to know if that was the issue.
If I get a new drive, I assume it's safe to clone this one into the new one?

I looked up the drive and it is a Seagate Barracuda ST3000DM001 3TB 64MB. Funny enough I paid $100 for it in 2012, and its updated version is still $100 on newegg today (ST3000DM008)

Yes, the drive is failing.
Get a new HD (or SSD) ASAP.
Use this one for scratch, or take out the magnets and play with them.

 

[UsandThem]

CrystalDiskInfo reports everything nominal except:

05 100 100 _36 000000000028 Reallocated Sectors Count
C5 100 100 __0 000000000048 Current Pending Sector Count
C6 100 100 __0 000000000048 Uncorrectable Sector Count

I don't really know how to interpret SMART values but the Acronis database says all 3 are critical and could be a sign of imminent failure. Sounds like I should replace the drive?

I moved it to another port and so far so good, but since the problem has been intermittent it's hard to know if that was the issue.
If I get a new drive, I assume it's safe to clone this one into the new one?

I looked up the drive and it is a Seagate Barracuda ST3000DM001 3TB 64MB. Funny enough I paid $100 for it in 2012, and its updated version is still $100 on newegg today (ST3000DM008)

Yes, the drive is failing.
Get a new HD (or SSD) ASAP.
Use this one for scratch, or take out the magnets and play with them.

Yup, it's a goner. SMART monitoring is nice, but a person should manually run a utility like CrystalDisk every 6 months or so to see if errors are beginning to show up. Many times by the time SMART actually throws a warning to you, your drive is dead/dying. At least you got a little warning with your drive appearing/disappearing so you could save any important files.

 

[tortillasoup]

Sometimes you can get relocated sectors if you overheat a drive from doing excessive data transfer without sufficient cooling. Aside from that, using tools like spinrite can sometimes add life to the drive if the bad sector was written erroneously.

 

[Elixer]

Sometimes you can get relocated sectors if you overheat a drive from doing excessive data transfer without sufficient cooling. Aside from that, using tools like spinrite can sometimes add life to the drive if the bad sector was written erroneously.

Max operating temps for a HD is around 55C-60C (131F-140F), and once you get to those temps, you will have more issues than just reallocated sectors.
Spinrite is destructive, you NEVER want to use it on a drive that you are trying to get data back from if you didn't clone the drive first.
Many times, that programs makes things worse, and then people send the drive to the data recovery people, and there is pretty much nothing they can do since it performed destructive operations on the drive already.

 

[tortillasoup]

Max operating temps for a HD is around 55C-60C (131F-140F), and once you get to those temps, you will have more issues than just reallocated sectors.
Spinrite is destructive, you NEVER want to use it on a drive that you are trying to get data back from if you didn't clone the drive first.
Many times, that programs makes things worse, and then people send the drive to the data recovery people, and there is pretty much nothing they can do since it performed destructive operations on the drive already.

Well I guess it depends on the failure mode. I've had drives ten years ago that @ 55C, they were writing junk data and ended up corrupting the MFT but in better scenarios just junk data blocks. After cooling off the drive, I would be able to fix the data structure and it was fine from there. Now, these newer drives, they reach 50C and they start writing garbage data. The really old drives from the year 2000 and earlier, never had an issue with them 'overheating'. One custom feature that I've noticed can't be enabled on newer seagate drives is the 'quiet' vs 'performance' mode. Quiet mode seems to make the drive run cooler and is likely much safer when you don't have active cooling for the drives.

As for spinrite being destructive, I think it depends on the failure mode. For me, I found it invaluable because I've had drives that had various bad sectors which were recovered well enough to then allow me to properly image the drive. The imaging software I would use typically got hung up on the bad sectors and it made imaging the drive impossible. Did spinrite possibly write garbage data in those sectors despite re-reading several times and averaging the values? Possibly but it was lost data anyway and since the rest of the drive had good data, being able to image the drive to a good drive, then running check disk (another destructive piece of software), I had a then usable file structure and or OS install.


I've yet to meet a person who actually has had success with a 'data recovery' specialist. I'm not saying it's impossible as I have done data recovery myself with certain software, but in all the stories and recommendations, yet to know anybody who has had real success. Just lots of money wasted and no results to speak of. I mean the idea that you could spend $1000 for "data recovery" and then have no results to speak of? That's insane.